Disney and NVIDIA Team Up on Artificial Intelligence for Making Better Movies
Walt Disney Co. (NYSE:DIS) presented at a conference last month its newly developed artificial-intelligence (AI) technology for tracking the facial reactions of theatergoers as they watch a movie. The tech could help the entertainment giant make even better movies and could also have other applications across its empire.
Source: Beth McKenna
Disney tapped an NVIDIA (NASDAQ:NVDA) graphics processing unit (GPU) to play a starring role in the development of its new tech.
Here's what you should know.
People in movie theater -- man, woman, boy, and girl in front, with others in rows behind them.
Disney's new factorized variational autoencoders (FVAE) tech falls within the branch of AI called deep learning, which aims to mimic human thought processes. In deep learning, an artificial neural network is trained how to think or make inferences, and then it's deployed where it makes inferences from new data, which could be images, speech, and so on. NVIDIA's GPUs have several applications, from high performance computing to accelerating the time it takes to train these neural networks and beyond. In this case, Disney Research used the graphics-chip maker's Tesla K40 GPU to evaluate the effectiveness of FVAE against several other AI models.
Disney's new AI tech has been trained to watch an audience of faces in a dark movie theater to track their reactions -- such as smiles, frowns, and signs of fright, which would be a desired reaction for scary scenes. The tech can track reactions on a very granular level, which should make it much better at gauging movie watchers' true feelings while watching a film than self-reporting methods such as interviews and questionnaires, which are notoriously unreliable.
The most obvious use of the tech is with test audiences. Disney could use the tech when testing movies before launch and make final edits based on audiences' reactions. The tech could also lead to "responsive storytelling," where the story changes depending upon an audience's reactions. Outside the movie realm, Disney could also use its tech at its theme parks to gauge visitors' reactions to various attractions and use the results to make changes aimed at improving visitors' enjoyment.
Potential privacy issues would have to be worked out. The use of the tech with movie test audiences shouldn't present an issue; Disney could disclose in advance that the tech was being used, and folks who had a problem with it just wouldn't participate. Things get trickier when it comes to general movie audiences and park visitors.
Middle-aged man and woman sitting in a theater and laughing hard, suggesting they're watching a funny scene in a movie. A few other people are visible in seats behind them.
Disney's movie business has been hitting it out of the park in recent years. In fiscal 2016, this segment's revenue jumped 28% year over year to $9.44 billion, accounting for 17% of the company's total revenue. Its operating income soared 37% year over year to $2.7 billion, accounting for 17.2% of total segment operating income. The movie business is more important to Disney's overall business than those figures suggest -- hit movies provide content for new theme-park attractions and for a plethora of licensed consumer products.
Fiscal 2017 is unwinding as a slower-growth year for Disney's movie business, as management expected. The company has churned out hits such as Rogue One: A Star Wars Story, Beauty and the Beast, and Guardians of the Galaxy Vol. 2. However, that hasn't been enough to jump the sky-high comparables bar. Fiscal 2016 opened with the release of the megahit Star Wars: The Force Awakens, followed by box-office champs Captain America: Civil War, Finding Dory, Zootopia, and The Jungle Book.
The good news for investors is that Disney CEO Bob Iger said going into fiscal 2017 that Disney's more recent strong growth dynamics will continue again in fiscal 2018.
The final wrap
Disney's new AI tech could help it makes movies that resonate even better with audiences, even if only used with test audiences. And even if this tech never sees the light of day -- or the dark of movie theaters -- beyond the research stage, it could act as a springboard for the company to develop other AI technologies. Anything Disney can do to help increase consumers' enjoyment of its products should help it increase its profits.
As for NVIDIA, it's obviously good news for the burgeoning AI player that Disney used one of its GPUs in developing its new AI tech. It seems likely that Disney -- which uses NVIDIA's GPUs for various other applications -- will continue to use NVIDIA's GPUs in its explorations of other deep-learning techs. NVIDIA's GPU-based deep-learning approach to AI is emerging as a favored approach to AI across a wide range of applications, from data centers to driverless cars.